激光数据线提取在matlab中使用霍夫变换
我有一组激光测距仪数据,将其转换为全局框架后,它有负数和浮点数据,例如xy坐标是:
x=[1.1 -2.2 3.45]
y=[0 4.67 -5.6]
如何使用matlab霍夫变换提取线?
看来输入必须是图像。
I have a set of laser range finder data, after transform it to global frame, it has negative and floating point data, for example the x-y coordinate are:
x=[1.1 -2.2 3.45]
y=[0 4.67 -5.6]
how to use matlab hough transform to extract line?
it seems the input must be an image.
如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱,如果其他用户或者作者回复了您的评论,将不能在第一时间通知您!
发布评论
评论(2)
您不需要霍夫变换。
相反,使用最小二乘法找到最佳线:
您可以在此处阅读有关最小二乘法的信息:
http://en.wikipedia.org/wiki/Linear_least_squares_%28mathematics%29
You don't need the Hough transform.
Instead, find the best line using the Least Squares method:
You can read about least squares here:
http://en.wikipedia.org/wiki/Linear_least_squares_%28mathematics%29
我目前面临类似的问题。您可以用来提取特征的方法之一是 RANSAC。
有了这个,您可以匹配一些线,然后从线的交点中提取特征。显然,这种方法对于大数据集效果更好。对于大多数激光测距仪,您可以获得数百个点。
I am currently facing a similar problem. One of the methods you could use to extract features is RANSAC.
With this you could match some lines and then extract features from the line intersections. Obviously this method is better with a large dataset. And with most Laser range finders you get several hundred points.